How Airlines Actually Set Flight Prices: Revenue Management Explained

May 6, 2026

You search for a flight on Monday and it costs £280. You search again on Thursday and it's £340. You ask your friend who booked the same route last week and they paid £195. None of this is accidental. Airline pricing is one of the most sophisticated yield management systems ever built, and understanding how it works changes how you buy tickets.

The Bucket System

Airlines don't think about individual seat prices. They think about booking classes. A single economy cabin on a flight from London Heathrow (LHR) to New York JFK might contain 180 seats, but those seats are divided into perhaps 10 or 12 distinct fare buckets. Each bucket has a code — Y, B, M, H, K, Q, and so on depending on the carrier — and each bucket has a price, conditions, and an inventory allocation.

The cheapest bucket might have 8 seats allocated. Once those sell, the airline opens the next bucket at a higher price. When that fills, the next one opens. This is why prices tend to rise as a departure date approaches: the cheap buckets empty first.

Revenue Management Systems

Airline pricing isn't done manually by humans sitting at spreadsheets. It's handled by Revenue Management Systems (RMS) — proprietary software that processes enormous amounts of data to optimise how many seats to sell in each bucket at each price point.

These systems ingest historical booking curves (how similar flights on similar dates have sold in the past), current booking pace (whether this particular flight is selling faster or slower than the model predicts), competitive pricing from other carriers, events at the destination, and elasticity models that estimate how price-sensitive demand is on a given route.

The goal of the RMS is not to fill every seat — it's to maximise total revenue. An airline would rather fly with 3 empty seats and an average fare of £350 than fill those seats at £80 each. This is why last-minute seats on popular routes can be extraordinarily expensive rather than cheap: the RMS has identified that the remaining demand is highly inelastic (business travellers, emergencies) and prices accordingly.

Overbooking as a Feature

Airlines routinely sell more tickets than they have seats. This isn't a bug — it's a deliberate revenue strategy. Historical no-show rates on any given route are well-modelled. If past data shows that 6% of passengers on LHR–JFK on a the Tuesday booking myth evening don't show up, an airline might oversell by 4–5%, accepting some risk of having to rebook passengers in exchange for capturing revenue from those 6% who paid but didn't fly.

When overbooking results in volunteers or involuntary denied boardings, the compensation rules vary by jurisdiction. EU Regulation 261/2004 mandates significant cash compensation for denied boarding within the EU. US DOT rules apply on domestic US routes. The airline has already priced this compensation risk into its overbooking model.

Why Prices Vary by Market

This is the part most travellers don't know. Airlines file different fares in different markets. A British Airways (BA) ticket from London to Chicago sold through the UK booking system may carry a different base fare than the same ticket sold through the US booking system or the German system. Currency exchange explains some of the variance, but airlines also use regional yield management that accounts for local demand patterns, purchasing power, and competitive dynamics.

In markets with strong local competition, airlines may suppress fares to avoid losing market share. In markets where they have little competition, the same itinerary might be priced higher. This is why using a service like RegionFare — which simultaneously checks what the same flight costs across dozens of national booking markets — can surface meaningful savings on international routes. The fare differences aren't random; they're the result of deliberate regional pricing strategy.

The Role of GDS and Distribution Costs

Airlines don't just sell tickets directly. A large percentage of sales flow through Global Distribution Systems (GDS) — technology platforms like Amadeus, Sabre, and Travelport that aggregate inventory and feed it to travel agencies and online booking sites. GDS fees add to the cost of distribution, which is why airlines have steadily pushed travellers toward direct booking through their own websites.

This creates a pricing wrinkle: some fares exist only in certain distribution channels. Web-only fares are sometimes cheaper than GDS-filed fares because the airline avoids the distribution fee. Conversely, some corporate fares negotiated through travel management companies are cheaper than anything available to the public.

Dynamic Pricing and Personalisation

Modern revenue management has introduced a further layer: dynamic offer pricing. Instead of a fixed menu of fare buckets, airlines increasingly generate personalised price offers based on the requester's profile. Booking history, frequent flyer tier, device type, time of day, and even browsing history can all theoretically feed into what price you see.

The extent to which this happens varies by carrier. There's a meaningful difference between offer optimisation (showing you the most relevant bundled fare) and outright price discrimination (charging you more because your browsing history suggests you're willing to pay). The latter remains controversial and largely unprovable from the outside.

What This Means for Buyers

Understanding RMS behaviour leads to a few practical principles. First, the cheapest seats go first — the fare you see today may not be there tomorrow, and waiting rarely rewards patience unless you're shopping in an active window of fare competition. Second, prices are not symmetric: a seat that rises from £200 to £320 will not necessarily fall back to £200 if the airline doesn't fill. Third, the same seat can legitimately cost different amounts depending on where and when you search.

The most durable strategy is to search broadly — across dates, departure airports, routing options, and markets — rather than anchoring on a single search result and treating it as the price. Revenue management systems are sophisticated, but they're also bounded by the laws of supply and demand. Demand drops in shoulder seasons, competition pushes fares down on contested routes, and sales happen when load forecasts miss. All of that creates genuine opportunities for travellers who know where to look.

Fare Classes Decoded

The letter codes airlines use for fare buckets — Y, B, M, H, Q, V, W, L, S, T, and others — have specific meanings that go beyond just the price. Each fare class carries a set of rules encoded in the Global Distribution System. These rules determine whether the ticket is refundable, whether it can be changed and at what fee, whether it earns frequent flyer miles and at what rate, whether the passenger can be upgraded, and whether the fare can be combined with other fare types on a journey.

Full-fare economy (typically Y or B class) is almost always refundable and changeable at low or no cost. It earns miles at the highest rate. It's what corporations buy when an employee's travel plans are fluid. It's also often twice the price of the lowest available fare on the same flight.

Discounted economy runs through H, K, Q, and L classes, roughly in descending order of flexibility and price. An H-class fare might be changeable with a fee; a Q-class fare might be non-changeable; an L-class fare might be fully non-refundable and non-endorsable. The restrictions increase as the price decreases.

Business class has its own parallel ladder — J (full), C (discounted business), D and I class represent further discounted tiers. When airlines run business class sales, they're typically opening up more inventory in C, D, or I class rather than reducing the J-class base fare.

Understanding fare class matters for frequent flyers because mileage accrual is typically class-dependent. A deeply discounted economy fare in L or V class may earn as few as 25–50% of miles compared to 100% for a Y-class ticket on the same flight. The headline price difference between fare classes sometimes underestimates the true cost when you factor in forfeited miles on a long-haul route.

How Airline Sales Actually Work

Airline sales are not price reductions in the conventional retail sense. They are controlled openings of lower fare class inventory on specific routes for specific travel windows. When British Airways announces a flash sale, what's actually happening is that the revenue management system has unlocked additional seats in a low-bucket fare class — say Q or V class — that were previously held back.

This has several implications. First, sales are not available for all dates: the inventory opening is selective, typically covering shoulder-season travel periods where demand forecasts show a surplus of unsold seats. Second, sales are not available for all routes: the airline opens low-class inventory where it has the most competitive pressure, not on routes where demand is already strong. Third, sales expire not just when the promotional period ends but when the unlocked inventory sells out — which on popular routes can be within hours of the sale going live.

The trigger for a sale is usually a load factor problem. If the RMS projects that a set of flights in eight weeks will be 15–20% underbooked relative to target, the system may automatically unlock lower fare classes to stimulate demand. This is why the connection between booking timing and price is not linear: a flight might be expensive in week one, cheap in week four when a sale fires, expensive again in week six when the sale inventory exhausts, and then very expensive in the final two weeks when business travellers fill the remaining seats.

Dynamic Pricing in Practice: A Real-World Example

To make the RMS mechanism concrete, consider a London Heathrow (LHR) to Barcelona (BCN) flight on a Tuesday in late September — a shoulder-season midweek route with moderate leisure demand. At 90 days out, the cheapest available fare might be £89 return in Q class with Vueling or Iberia. At 60 days, if the flight is selling to plan, the price might hold at £89. At 45 days, if a competitor runs a flash sale on the same route, Iberia's RMS might respond by opening additional Q-class seats — temporarily dropping the displayed fare. At 30 days, if the flight is now close to its projected load target, those Q-class seats close and the displayed price jumps to £135 in H class. At 14 days, business travellers fill the final seats at £220 in B class.

None of those prices are wrong. Each reflects the airline's current best estimate of how to maximise total revenue on that specific flight given how demand has evolved. The traveller who booked at 45 days during the competitor-triggered dip got the best outcome — not because they followed a rule, but because they happened to search at the right moment.

What You Can Actually Control

The honest conclusion from understanding airline pricing is that you cannot reliably predict or time the exact lowest fare. But you can improve your expected outcomes by acting on a few structural realities.

Search on multiple dates simultaneously. The price difference between a Tuesday departure and a Friday departure on the same week is often larger than the discount from any timing trick. Midweek travel consistently produces lower fares on leisure-heavy routes because business travellers cluster around Monday morning and Friday afternoon.

Use price history to calibrate expectations. Several tools (Google Flights' price graph, Hopper's predictive model, Kayak's price trend indicator) give a sense of whether the current price is high or low relative to recent history for that route and season. This doesn't tell you whether prices will rise or fall, but it tells you whether you're looking at a real deal or a normal price.

Don't anchor on the first price you see. The same itinerary searched across different booking markets — UK, German, Italian, Japanese, Korean — can show materially different prices even for the same seat on the same flight. The fare is filed separately in each market's GDS inventory, and regional demand patterns mean that some markets get access to lower-class fares more readily than others. This cross-market price variation is the most consistently actionable opportunity in airline pricing — and the one that the average traveller never thinks to check.